Invalid memory address or nil pointer dereference on cluster restart

Server version is 4.6.3 GA (but it happens regardless of the server)
Go SDK codes are
https://github.com/brett19/gocbcore/
branch:v7
commit:5f7997accbcac0ffb0d456d8a13582172dfc1e1b

https://github.com/brett19/gocb
branch:master
commit:2eb0acf712d18731419004e82c57cd80d9914db4

When all servers are restarting one by one with 5 seconds sleep,
As soon as the first server stops (systemctl stop couchbase-server), then panic shows up

[88.65 DEBUG] (SSHCommand execute:75) Running service couchbase-server stop on 172.23.98.167
[88.99 INFO] (SDKD log:137) panic: runtime error: invalid memory address or nil pointer dereference
[88.99 INFO] (SDKD log:137) [signal SIGSEGV: segmentation violation code=0x1 addr=0x10 pc=0x6af14f]
[88.99 INFO] (SDKD log:137)
[88.99 INFO] (SDKD log:137) goroutine 98 [running]:
[89.00 INFO] (SDKD log:137) github.com/couchbase/gocb.(*Bucket).lookupIn.func1(0x0, 0x8e2fa0, 0x93d800)
[89.00 INFO] (SDKD log:137) /root/jenkins/workspace/sdk-go-situational-release/go-sdk-situational-watson-4.6.3/src/github.com/couchbase/gocb/bucket_subdoc.go:171 +0x9f
[89.00 INFO] (SDKD log:137) Starting SDKD on port 842350556864
[89.00 INFO] (SDKD log:137) gopkg.in/couchbase/gocbcore%2ev7.(*Agent).LookupInEx.func1(0x0, 0xc4208e65a0, 0x8e2fa0, 0x93d800)
[89.00 INFO] (SDKD log:137) /root/jenkins/workspace/sdk-go-situational-release/go-sdk-situational-watson-4.6.3/src/gopkg.in/couchbase/gocbcore.v7/agentops_subdoc.go:432 +0x4a0

Expected : no panic and application should continue